Risk of postnatal depression or suicide after in vitro fertilisation treatment: a nationwide case– control study Josefin Vikström Eckevall, Gunilla Sydsjö, Mats Hammar, Marie Bladh and Ann Josefsson Journal Article N.B.: When citing this work, cite the original article. Original Publication: Josefin Vikström Eckevall, Gunilla Sydsjö, Mats Hammar, Marie Bladh and Ann Josefsson, Risk of postnatal depression or suicide after in vitro fertilisation treatment: a nationwide case– control study, British Journal of Obstetrics and Gynecology, 2016 http://dx.doi.org/10.1111/1471-0528.13788 Copyright: Wiley: 12 months http://eu.wiley.com/WileyCDA/ Postprint available at: Linköping University Electronic Press http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-132763 Risk of postnatal depression or suicide after in vitro fertilization treatment: A nationwide case-control study Josefin Vikström MD, Gunilla Sydsjö PhD, Mats Hammar MD PhD, Marie Bladh MA, Ann Josefsson MD PhD Affiliations for all authors: Obstetrics and Gynecology, Department of Clinical and Experimental Medicine, Faculty of Health Sciences, Linköping University, Sweden Department of Obstetrics and Gynecology in Linköping, County Council of Östergötland, Linköping, Sweden. Corresponding Author Josefin Vikström Obstetrics and Gynecology, Department of Clinical and Experimental Medicine, Faculty of Health Sciences, Linköping University, Sweden Department of Obstetrics and Gynecology in Linköping, County Council of Östergötland, Linköping, Sweden. E-mail: [email protected], Telephone: +46703320527 Running title: Risk of postnatal depression after IVF treatment 1 Abstract Objective: To examine if women who undergo in vitro fertilization (IVF) treatment are at greater risk of postnatal suicide or postnatal depression (PND) requiring psychiatric care compared to women who conceive spontaneously. Design: Case-control study using data from national registers. Setting: Sweden during the period 2003-2009. Population: Cases were 3532 primiparous women who had given birth following IVF treatment. An aged-matched control group of 8553 mothers was randomly selected from the medical birth register. Methods: Logistic regression analyses were performed with PND as outcome and known risk factors of PND as well as IVF/spontaneous birth as covariates. Main outcome measures: PND defined as ICD-10 diagnoses F32-39 within 12 months after childbirth. Results: Initial analyses showed that PND was more common in the control group than the IVF group (0.8% vs. 0.4%; p=0.04). However, these differences disappeared when confounding factors were controlled for. A history of any psychiatric illness (p=.000; OR=25.5; 95 % CI=11.7-55.5), any previous affective disorder (p=,000; OR=26.0; 95%CI=10.5-64.0) or specifically a personality disorder (p=.028; OR=3.8; 95 % CI=1.2- 12.7) increased the risk of PND. No woman in either group committed suicide during the first year after childbirth. Conclusions: While mothers who receive IVF treatment are not at increased risk of PND, the risk is increased among mothers with a history of mental illness. Keywords: IVF, infertility treatment, postnatal depression, postnatal mental illness 2 Tweetable abstract: A Swedish study on 3532 women showed that IVF treatment does not increase the risk of postnatal depression. 3 Introduction It is generally reported that about 13% of women in developed countries suffer from postnatal depression (PND) 1, 2. However, studies have shown that the prevalence of PND cases that require psychiatric care is 0.3% 3 and that the prevalence of hospitalizations due to PND is 0.06%-0.26% 4. Apart from the negative effects on the woman’s health, PND can also have a negative impact on mother-child interaction and child development 5-7 . Risk factors of PND include previous psychiatric illness 8-10, specifically depression 9-10, antenatal depression and anxiety 8-10 , stressful life events 8-10 , lack of social support 8-10, a negative attitude towards pregnancy 8-9, neuroticism 9, low self-esteem , poor marital relationship, low socioeconomic status, being single, obstetrical stressors, and a difficult 9 temperament in the infant . However, the above mentioned risk factors cannot fully explain the variance in PND incidence 11. Gleicher (2007) argued that the pathogenesis of PND could be autoimmune based on similarities between PND and well-known autoimmune disorders 12. The rationale behind this was that pregnancy reveals underlying dispositions and cause recurrences for both autoimmune and affective disorders, both exacerbate during late pregnancy or the postpartum period and both have shared risk factors such as primiparity and childbirth by caesarian section 12. It has also been hypothesized that hormone exposure, specifically disruptions in the hypothalamus-pituitary-adrenal (HPA) axis, during pregnancy might be a factor in the PND pathogenesis 11. The theory behind this is that stress, which is a well-known cause of depression, is closely linked to the function of the HPA-axis and evidence of HPA-axis 4 dysregulation during the postpartum period has been found 11. Since infertility and its treatment causes levels of stress to rise during and after treatment 12, 13 one might hypothesize that the risk of HPA-axis dysregulation, and thus PND, would be increased in these women. Notably, An et al (2013) found a positive association between cortisol levels and state anxiety in women treated with in vitro fertilization (IVF) 14. However, most studies, with a few exceptions 16-18, have found the risk of developing postnatal depressive symptoms or PND to be equal when those who have conceived using assisted reproductive technologies (ART) are compared to those who have conceived spontaneously 19-23. However, Ross et al (2011) reported that most studies have failed to control for known risk factors of PND 20. In general, women who undergo infertility treatment have fewer risk factors of PND. They have a higher socioeconomic status, enjoy relationships of a better quality with a longer duration and are more psychologically well- adjusted than those who conceive naturally 24-28. Therefore, Ross et al. (2011) speculated that future studies controlling for these factors may in fact find the risk of PND to be increased among women who have undergone infertility treatment. Also, most previous studies have used self-report depression screening instruments rather than diagnostic criteria to identify cases 20. Since an increasing number of couples use IVF treatment to conceive 29 it is important to determine if this treatment increases the risk of PND, which might have grave consequences on the health of both mother and child 5-7. We hypothesized that women who receive IVF treatment are at increased risk of PND since previous studies have found the risk among these women to be equal to those who have conceived spontaneously while failing to control for possible confounding factors. The aim of the present study was to examine if women who undergo IVF treatment are at greater risk of 5 postnatal suicide or PND requiring psychiatric care compared to those who conceive spontaneously. Methods Population All primiparous women born from January 1st 1973 to December 31st 1983 who were included in the Swedish IVF register from January 1st 2003 to December 31st 2009 were included in the study (n=3532). This time interval was chosen so that as much information as possible would be available from all registers for all participants. A control group of women (n=8553) who had conceived spontaneously and subsequently given birth were, using SPSS sample procedure, randomly selected from the Medical Birth Register (MBR). In order to increase the power of the study, the control group chosen was more than twice the size of the IVF group. The groups were matched for age. Registers The Medical Birth Register (MBR) The MBR includes nearly all pregnancies that have resulted in births in Sweden since 1973 and contains information about the pregnancy, delivery and antenatal health of the child. Due to reporting errors, one to three percent of deliveries are missing from the register each year. The register is based on the medical charts from antenatal, obstetric as well as neonatal care 30 and has been thoroughly described by Harlow et al. (2007) 31. Using the personal identification number (PIN) of the women, information about the height, weight, parity, previous miscarriages, previous experience of delivering a stillborn child or death of a child during the first month, gestational week at childbirth as well as mode of childbirth was collected. Birth weight was categorized into extremely low birth weight 6 (<1000g, ELBW), very low birth weight (1000g-1499g, VLBW), low birth weight (1500- 2499g, LBW) and normal birth weight (>2500g). Gestational age was divided into <28 weeks (extremely preterm), 28-31 weeks (very preterm), 32-36 weeks (preterm), 37-40 weeks (term) and >40 weeks (post term). The National Patient Register (NPR) The NPR contains information about all patients who have been discharged from hospital or treated in outpatient care. It does not contain diagnoses received in general practice. This register has been thoroughly described by Harlow et al (2007) 31. Pre- and postnatal psychiatric diagnoses were collected from the NPR. The diagnoses were limited to the past five years before, and one year after, childbirth and were chosen from the fifth chapter of the ICD-10, containing the mental and behavioural disorders. The DSM-IV definition of postnatal depression is